Devstral 2

MistralMCP Tool Call

mistral/devstral-2512Open Weights

Mistral's coding-agent model for repository work, terminal tasks, and software fixes

Benchmark Pricing$0.40 / 1M in$2.00 / 1M out

Context: 262,144 tokens•Max Output: 262,144 tokens•Knowledge: 2025-12•Released: 2025-12-09

TECHNICAL SPECIFICATIONS

Context Window262,144262.1K tokens

Max Output262,144262.1K tokens

LicenseOpen SourceSelf-hostable

Input Modalities

Text

Function / Tool CallingSupported (MCP Ready)

Structured OutputsGuaranteed Schema

MODEL CONTEXT PROTOCOL (MCP) READINESS

✓ Devstral 2 natively supports OpenAI / Anthropic format tool calling and is fully compatible with MCP servers and agent loops.

MCP Context Window Recommendations

With a context capacity of 262,144 tokens, this model can handle moderately large MCP tool definitions and multi-turn conversation logs comfortably.

STANDARDIZED BENCHMARKS & EVALS

Artificial Analysis Coding Indexindex

23.7

SciCodepercent correct

33.1

Terminal-Bench Hardsuccess rate

18.9

API CALL INVOCATION EXAMPLE

// Example: Calling Devstral 2 via OpenAI SDK
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.MISTRAL_API_KEY,
});

const response = await client.chat.completions.create({
  model: "mistral/devstral-2512",
  messages: [{ role: "user", content: "Hello! Can you inspect MCP tools?" }],
});

console.log(response.choices[0].message.content);